QT dispersion from body surface ECG does not reflect the spatial dispersion of ventricular repolarization in sheep

The correlation between the QT dispersion on body surface ECG and the dispe rsion in ventricular repolarization from the cardiac surface was studied in six sheep anesthetized with pentobarbital. The standard 12-lead body surfa ce ECG and multiple ventricular epicardial ECGs were simultaneously recorde d. The activation-recovery interval (ARI) was measured from the unipolar ep icardial ECGs. The pooled QT dispersion from the six animals was significan tly smaller than the pooled ARI dispersion (22.7 +/- 2.6 vs 33.0 +/- 6.9 ms , P < 0.01). There was no correlation between the QT and ARI dispersion. Th e unipolar epicardial ECGs were then converted into bipolar ECGs and epicar dial QT intervals were subsequently acquired from these ECGs. The average v alue of epicardial QT dispersion from the six animals was similar to that o f body surface EGG, but was less than the ARI dispersion (27.5 +/- 6.8 vs 3 3.0 +/- 6.9, P < 0.01). A good correlation between the epicardial QT disper sion and ARI dispersion was identified (r = 0.84, P < 0.05). In addition, a prolongation in ventricular repolarization, induced by an increase in coro nary flow, elicited a pooled ARI dispersion of 62.3 +/- 6.2 ms (n = 6), whi ch was larger than the simultaneously recorded body surface QT dispersion ( 28.3 +/- 9.8 ms, n = 6, P < 0.01). No correlation between the ARI and QT di spersion was found in the presence of the prolonged ventricular repolarizat ion. In conclusion, QT dispersion from a 12-lead body surface ECG seems to underestimate the spatial dispersion of ventricular repolarization acquired from sheep epicardium.